Hey. I am referencing my R console output as follows:

> y <- matrix(rnorm(5*2),5,2)
> design <- matrix(c(1,1),2,1)
> > # First set of 3 genes are genuinely differentially expressed
> index1 <- 1:5
> y[index1,2] <- y[index1,2]+1
> > # Second set of 2 genes are not DE
> index2 <- 4:5
> > c=camera(y, list(set1=index1,set2=index2), design)
> y
         [,1]                    [,2]

[1,] -1.0451966  -1.5088467
[2,] 0.5473909  -0.1203236
[3,] 0.4864850  0.8291973
[4,] 0.9755390  -1.3448109
[5,] -1.4054274  0.3586121
> index1 [1] 1 2 3 4 5
> index2 [1] 4 5
> c
          NGenes Correlation Direction PValue        FDR
set2     2                -1.0           Down 0.6768028 0.6768028
et1       5                -0.2             Up      NaN          NaN

Request to please help me understand a few points. I understand that 'y', my input contains two samples of 5 genes each. and my index1 and index2 are shown above. I wanted to know what does this output signify looking at the Pvalues and why do we need to specify 2 samples in 'y'. As far as I understand, for set 2 we are comparing the genes 4 and 5 (in the set) to the genes 1, 2 and 3 (not in the set).
Thanks for your help.

Regards,
Akarshan Puri

The matrix contains all the expression data for your experiment. Usually this contains many more than 2 columns. y must contain replicate samples because you can't use camera unless you can test for differential expression, and you can't do a test for differential expression unless you have replication.

I suspect from your post that you might not understand what camera is doing. camera is not testing for differential expression between genes. The place to start would be to read the camera publication:

 http://nar.oxfordjournals.org/content/40/17/e133